J
jho1986
New member
Whats the difference between recording at 16 bits or 24 bits? Also... is there a bigger difference between 44.1 khz and 96 khz or between 16 bits and 24 bits?
Kevin Deschwazi
Well-known member
As far as bit depth goes I do notice a difference at 24 as opposed to 16. Not so much in the sound quality but in that I tend to get more headroom, it's definitely easier to work with.
I notice no difference whatsoever recording above 44.1 khz. In fact Dan Lavry wrote a white paper on sampling rate a while ago and he concluded that recording at too high a sample rate was actually detrimental to the audio. I tinkered with 48khz for a while 'just in case' but went back down.
So there you go, I now record at 24/44.1.
There are quite a few threads on this subject if you try a search.
I notice no difference whatsoever recording above 44.1 khz. In fact Dan Lavry wrote a white paper on sampling rate a while ago and he concluded that recording at too high a sample rate was actually detrimental to the audio. I tinkered with 48khz for a while 'just in case' but went back down.
So there you go, I now record at 24/44.1.
There are quite a few threads on this subject if you try a search.
Bass Master "K"
New member
Kevin is right about doing the search. You will find loads of threads and there are a few really big ones with loads of posts on them. You will find all sorts of opinions on the matter. I think with the 24 v 16 issue, most will agree that 24 is easier to work with, but many find 16 works just as well for what they do. 16 bit is easier on your computer if you have an older model.
I haven't read the paper that Kevin mentions about not recording at too high of a rate, but the last article I read about it, and of course I can't remember who the author was, but it was someone in the industry. He said that recording a rock band or the like, 96 was unneccisary, but that he did a comparative situation with an orchestra and that at 96 he could 'clearly' hear the high end harmonics coming through that weren't at 44.1. His opinion was any insturments that have high end harmonic importance could benifit from 96.
What I always wondered was why 96 and not 88.2 which should math down easier? Ahhhh...It almost makes me miss the days when all I did was hit record on my old analog system.
I haven't read the paper that Kevin mentions about not recording at too high of a rate, but the last article I read about it, and of course I can't remember who the author was, but it was someone in the industry. He said that recording a rock band or the like, 96 was unneccisary, but that he did a comparative situation with an orchestra and that at 96 he could 'clearly' hear the high end harmonics coming through that weren't at 44.1. His opinion was any insturments that have high end harmonic importance could benifit from 96.
What I always wondered was why 96 and not 88.2 which should math down easier? Ahhhh...It almost makes me miss the days when all I did was hit record on my old analog system.
Kevin Deschwazi
Well-known member
Ah yes I was intending to come back and edit my post to add the link. Thanks for the reminder!Bass Master "K" said:
http://www.lavryengineering.com/documents/Sampling_Theory.pdf
....there you go.
Bass Master "K"
New member
Hey Kevin, thanks for posting that article up. The math was over my head, but it was an interesting read. I've been out of the loop for a while...how has this article been recieved? Are there those who don't agree with his theory or has it been widely accepted?
Kevin Deschwazi
Well-known member
The maths goes over my head too. 
I've not read any criticism of the paper, here on homerec or elsewhere but of course that's not to say there isn't any at all. I guess there aren't many people around with enough in-depth knowledge of the subject to seriously critique a white paper on sampling rate written by Dan Lavry, so people generally seem to be accepting it.
Maybe try searching a couple of the forums where the big name guys hang out? It might be interesting to see what (if anything) those guys have to say about it?
I've not read any criticism of the paper, here on homerec or elsewhere but of course that's not to say there isn't any at all. I guess there aren't many people around with enough in-depth knowledge of the subject to seriously critique a white paper on sampling rate written by Dan Lavry, so people generally seem to be accepting it.
Maybe try searching a couple of the forums where the big name guys hang out? It might be interesting to see what (if anything) those guys have to say about it?
Chris Shaeffer
Peavey ROCKS!!!!!!!!!!!!!
In general, record at the highest bit rate you can with either the target sample rate (44.1 for CD, 48 for video) or twice the sample rate. DON'T use 24/96 for CD just because it the highest quality. The math it takes to downsample 96 to 44.1 is too wicked... 88.2 is easy: just throw every other sample out.
In practice, unless someone is paying you to abuse your harddrive... record at 24/44.1. In most "prosumer" level gear you don't get enough (if any) benefit to justify the HUGE loss in track count and processing power with higher sample rates. And, as mentioned, the higher rates can in sound worse- the clocks and convertors don't necessarily do well with the faster sample rates.
16 can sound *great*. On the same hardware, though, 24 bit will sound better with less effort. I only use 16 with student projects and when I'm doing some random experiment. Even when I'm just messing around with an arrangement I tend to go with 24 bit just because I might end up using some of that stuff in the final version.
Take care,
Chris
In practice, unless someone is paying you to abuse your harddrive... record at 24/44.1. In most "prosumer" level gear you don't get enough (if any) benefit to justify the HUGE loss in track count and processing power with higher sample rates. And, as mentioned, the higher rates can in sound worse- the clocks and convertors don't necessarily do well with the faster sample rates.
16 can sound *great*. On the same hardware, though, 24 bit will sound better with less effort. I only use 16 with student projects and when I'm doing some random experiment. Even when I'm just messing around with an arrangement I tend to go with 24 bit just because I might end up using some of that stuff in the final version.
Take care,
Chris
Chris Shaeffer
Peavey ROCKS!!!!!!!!!!!!!
Oh. To answer your question directy:
The difference between 16 and 24 bit can be thought of as volume resolution. The same scale of volume for each sample is divided up into 16 or 24 bit words. I don't recall the exacty figure, but 24 bit isn't just half again as good as 16- its more like 8 TIMES the resolution as 16. Even 24 bit was regarded as overkill a while ago- 20 bit is fine and I use an ADAT XT-20 as a convertor for extra inputs on my 24 bit rig and it sounds just as good.
With the extra resolution in volume, 24 bit recordings are harder to clip and don't loose clarity when the volume gets lower. With 16 bit you can sometimes hear the quality of the sound degrading as the volume drops so you tend to record everything as hot as possible. But then it clips fairly easily.... so 24 bit is just way easier even if it doesn't sound noticeably better than well done 16 bit.
-C
The difference between 16 and 24 bit can be thought of as volume resolution. The same scale of volume for each sample is divided up into 16 or 24 bit words. I don't recall the exacty figure, but 24 bit isn't just half again as good as 16- its more like 8 TIMES the resolution as 16. Even 24 bit was regarded as overkill a while ago- 20 bit is fine and I use an ADAT XT-20 as a convertor for extra inputs on my 24 bit rig and it sounds just as good.
With the extra resolution in volume, 24 bit recordings are harder to clip and don't loose clarity when the volume gets lower. With 16 bit you can sometimes hear the quality of the sound degrading as the volume drops so you tend to record everything as hot as possible. But then it clips fairly easily.... so 24 bit is just way easier even if it doesn't sound noticeably better than well done 16 bit.
-C
C
carlosba
New member
Nah 24 bit / 192Khz
nah 24bit/192khz is what i use and the quality comes out professional... same thing PROTOOLS Flag Ship uses .. which is a $10,000 dollar hardware mixer.... trust me my producer told and he said theres is a big diffrence in quality its will sound crystal clear with a punch to it
nah 24bit/192khz is what i use and the quality comes out professional... same thing PROTOOLS Flag Ship uses .. which is a $10,000 dollar hardware mixer.... trust me my producer told and he said theres is a big diffrence in quality its will sound crystal clear with a punch to it
C
carlosba
New member
Forgot
its no regualer protools too.. if your tthinking? i'm talking about ProoTools/HD hardware... so the quality kills... if your asking how to get the same quality 24bit/192khz and not wasting $10000 dollars then i suggest buy EMU 1820M and do research if you don't bilieve... the only bad thing is that it has 2 mic in with 48 phantom power already in the soundcard... the reall protools comes with 24 mik ins... so it worth the buy if you just need it to record voice.. read the bottom if you might ask yourself what i reffer VVVVVV
its no regualer protools too.. if your tthinking? i'm talking about ProoTools/HD hardware... so the quality kills... if your asking how to get the same quality 24bit/192khz and not wasting $10000 dollars then i suggest buy EMU 1820M and do research if you don't bilieve... the only bad thing is that it has 2 mic in with 48 phantom power already in the soundcard... the reall protools comes with 24 mik ins... so it worth the buy if you just need it to record voice.. read the bottom if you might ask yourself what i reffer VVVVVV
Chris Shaeffer
Peavey ROCKS!!!!!!!!!!!!!
Yeah, but is it WARM and punchy? Or just punchy?
Sorry....couldn't resist.
Sorry....couldn't resist.
C
carlosba
New member
Amagine
Just imagine being in the studio if you haven't went... matter fact the music you listen to... either rocking roll or rap there voice is crystal clear... the same quality they record ... i gone to many studios and they use it... the EMU 1820m cost $499.99 and it comes with wavelab lie cubase.. and all the top of the line programs but not the real ones... they are real of course but its not the full fversion its just lite.. at the same time its no demo cause there is no limits
Just imagine being in the studio if you haven't went... matter fact the music you listen to... either rocking roll or rap there voice is crystal clear... the same quality they record ... i gone to many studios and they use it... the EMU 1820m cost $499.99 and it comes with wavelab lie cubase.. and all the top of the line programs but not the real ones... they are real of course but its not the full fversion its just lite.. at the same time its no demo cause there is no limits
C
carlosba
New member
for got again
the emu doesn't just record voice with two XRL mics ins in the back of the hardway it comes a dj pnono in to produce djaying and 3 L/R guitar ins and what ever you can put on good for home recording//// the go to www.emu.com and check it out it talks about it
the emu doesn't just record voice with two XRL mics ins in the back of the hardway it comes a dj pnono in to produce djaying and 3 L/R guitar ins and what ever you can put on good for home recording//// the go to www.emu.com and check it out it talks about it
Chris Shaeffer
Peavey ROCKS!!!!!!!!!!!!!
Hmm... I'm not saying you're wrong (because you aren't) but I've been in quite a few studios myself and many of them record 24/44.1 and 24/48 and it *still* sounds crystal clear. Have you done a side by side comparison of the 44.1 vs 88.2, 96, or 192 files?
Welcome aboard, by the way, carlosba!
Welcome aboard, by the way, carlosba!
warble
New member
Regarding 24bit/192khz.....well, this has been argued to death around here and all over the net. My thought is why use up all that extra storage space for what sonically is not going to be that different? I've haven't read about masses of people around here proclaiming that 192khz is the only way to go.
I've heard crystal clear recordings done in 24/44.1 (including my own) and I feel that the quality is fine. I've even tried some comparisons myself....I wasn't blown away by any sonic differences. Tried 88.2, tried 96...notta. I can't imagine that stepping up to 192 would rock my world.
carlosba, have you done comparisons, or is your opinion just based on what your producer told you? I certainly don't mean any of this an insult to you, just want to raise the questions and hear some concrete answers. The people that come here and aren't quite sure of what all the technology means, need to be informed the best possible way.
I've heard crystal clear recordings done in 24/44.1 (including my own) and I feel that the quality is fine. I've even tried some comparisons myself....I wasn't blown away by any sonic differences. Tried 88.2, tried 96...notta. I can't imagine that stepping up to 192 would rock my world.
carlosba, have you done comparisons, or is your opinion just based on what your producer told you? I certainly don't mean any of this an insult to you, just want to raise the questions and hear some concrete answers. The people that come here and aren't quite sure of what all the technology means, need to be informed the best possible way.
jpw23
New member
Chris Shaeffer said:
I thought that everything ended up as 16/44.1 at mixdown......is there a reason to record @ 24 if it gets dithered down to 16?
X
xstatic
New member
This is what I tell others. Record at 24 bit 44.1khz if you can, 16/44 if you can't. If your setup will easily accomadate you recording at 88, 96 or whatever higher khz without you having to make sacrifices, go for it. My personal belief is that the higher rates do probably sound a little better, but not what some people may say. There is no "magic" there that a 24/44 recording would not have. I also believe that it takes a much better signal chain and environment to really take advantage of the minor benefits that something like 24/192 would have.
Chris Shaeffer
Peavey ROCKS!!!!!!!!!!!!!
Yeah... it sounds better.
The higher quality source material you have for the final 16/44.1 dithering the better. I keep my mixs and masters at 24/44.1, by the way,for archiving and convert them to 16/44.1 or mp3 from there.
The thing that isn't obvious is that dithering only applies to bit rate reduction and not sample rate reduction. There is no clean way to smooth out sample rate change but dithering actually does make bit rate reductions good enough that recording at 24 sounds better even when the final product ends up at 16.
Take care,
Chris
The higher quality source material you have for the final 16/44.1 dithering the better. I keep my mixs and masters at 24/44.1, by the way,for archiving and convert them to 16/44.1 or mp3 from there.
The thing that isn't obvious is that dithering only applies to bit rate reduction and not sample rate reduction. There is no clean way to smooth out sample rate change but dithering actually does make bit rate reductions good enough that recording at 24 sounds better even when the final product ends up at 16.
Take care,
Chris
Farview
Well-known member
If your producer had to tell you there is a big difference, it can't be that big. 99% of the music in your CD collection was done at 44.1 or 48k. Anything that is more than a couple years old had to be, because there was no choice.carlosba said:
noisewreck
New member
I'm using 32/48. I find that 32bit gives me enough headroom, and figure there'll be less math errors during mixdown. 48kHz because that's what my synth likes to output at.